Hello, I am evaluating different clustering algo's from mahout : kMeans, Dirichlet, Mean-shift & Latent Dirichlet Allocation
I want to valiate accuracy. https://cwiki.apache.org/MAHOUT/clusteringyourdata.html metions about http://www.lucidimagination.com/search/document/dab8c1f3c3addcfe/validating_clustering_outputwhich seems broken. Are there classes / mechanisms in mahout (ClusterEvaluator) which could be used from command line which are not in mahout 0.7 ? http://comments.gmane.org/gmane.comp.apache.mahout.user/12953 Could you please point some documents which demonstrate how to validate clusters ? Thanks, Rajesh
